This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TPA3255: Transformer + LC architecture

Part Number: TPA3255

I am looking at interfacing a class D power amplifier with a high voltage transducer in an underwater acoustic application. Is there advantages or disadvantages of having the transformer at the output of the power amplifier and then the LC filter, or flipped where the LC is on the output of the PA? Any thoughts would be great.

Thanks

-David

  • Hi David,

    The only time I have seen the implementation of a transformer on the output of our devices, it was placed after the LC filter.

    I assume you are planning on using BTL mode?

    Regards,
    Robert
  • Hey David!

    How's everything! Good to hear from you!

    Yes, you can use the device with a transformer, but you need to have the LC filter before transformer. Most transformers will have higher losses with high frequency switching of the Class-D amplifier (450-600kHz). In addition, you may also need to insert a high pass filter before the Class-D amplifier to prevent saturation of the transformer at low frequencies (depending on your transformer design, size and core).

    I have an application note on some testing we did with transformers for audio applications that I'll send you, too.

    Let us know how we can help!

    Thanks,
    Brian
  • Hey

    Thanks for the info. The reason why I ask is that the current architecture that we have utilizes the IRS2092 in a differential setup (BTL) it has an LC filter at the output of the FETs and then a transformer to step up the voltages to several hundred voltages due to the ceramic transducer ring. The LC filter and transformer are extremely large. The audio signal is generated via a DAC from a DSP. The audio signal ranges from low 100's of HZ up to 20k to create tonal and broadband acoustic signatures. The transformer itself is multi-taped and connected through many relays to be able to switch between different frequency bands.

    The thing that we are trying to do is to decrease the solution size as the filter and transformer stages takes up a good amount of space.

    My understanding is that a high operating frequency will allow the magnetics to become smaller, but as you mentioned, the losses could potentially be worse and outweigh the size benefit. So by putting it at the output of the FETs could be possible, before the LC filter. I didn't know if that kind of implementation is acceptable. We are still trying to understand the original design and it's been in one of our legacy products for quite some time.

    If you could send me the app note and any other helpful material, that would be great!

    Thanks

  • Hi David,

    Did you get a chance to look at the app note? Let me know if you have any more questions. The best bet may be getting the TPA3255EVM and checking the power losses with your transformer.

    FYI, the TPA3255 also supports 2-inductor, Pre-Filter Parallel Bridge-Tied Load (PBTL) mode, which reduces the number of inductors from four to two. Alternatively, you can also shutdown the C/D channel if you are not using it.

    Thanks,
    Brian